the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 18 Aug 2026
High centennial-scale variability in an East Antarctic snow accumulation record spanning the last millennium
Abstract. Mass balance estimates of the East Antarctic ice sheet over the satellite era have a high degree of uncertainty, yet are of such significance that they can partially or even entirely offset the ice loss at the margins of the Antarctic Ice Sheet that contributes to global sea level rise. Therefore, it is imperative to improve our understanding of snowfall variability across time and space in East Antarctica. Here we present a new millennial-scale, annual snow accumulation history from the Mount Brown South (MBS) ice core, a semi-coastal record from 2,084 m elevation on the boundary of Wilhelm II and Princess Elizabeth Land in East Antarctica. We used a range of methods to test the best derivation of annual accumulation rates, and to ensure best practice in determining both density correction and layer thinning with depth. The MBS annual accumulation record spans 873–2017 CE, and displays a high degree of variability across time scales. The long-term mean snow accumulation rate at MBS is 0.276 ± 0.08 (1 standard deviation) m yr−1 ice equivalent, which is almost 10 % lower than the mean accumulation rate in the satellite era (1979–2017) of 0.302 ± 0.08 m yr−1 ice equivalent. While mean annual accumulation during the satellite era is within the range of natural variability, the full record exhibits substantial decadal to centennial-scale variability, including two notable low accumulation periods between CE 1127–1321 (195 years) and 1754–1879 (126 years), where accumulation was 12 % lower than the long-term mean, and 20 % lower than satellite era mean. While an ice core record is a point-based rather than spatially resolved estimate of annual accumulation, these results raise implications for future mass balance assessments in this region that assume the satellite era represents the natural range of variability.
- Preprint
(4856 KB) - Metadata XML
-
Supplement
(246 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-4880', Anonymous Referee #1, 19 Aug 2026
-
RC2: 'Comment on egusphere-2026-4880', Anonymous Referee #2, 08 Oct 2026
Review of “High centennial-scale variability in an East Antarctic snow accumulation record spanning the last millennium” by Nilssen et al.
The paper present and discuss snow accumulation data from a 120 m deep ice core drilled from a “semi-coastal” site in the Indian Ocean sector, in East Antarctica. The Mount Brown South (MBS) ice core covers about 1000 years and is from an area where not much data exists from before. Many challenges remain for estimating the Antarctic ice sheet surface mass balance (SMB), which represents a major uncertainty in predictions of future sea-level rise. Validating continental scale studies is hampered by the sparse distribution of in situ data and there are still large error bars on the estimates that do exist. There is generally a lack of high-resolution ice core data from Antarctica and therefore this drill site is particularly interesting. The ice core site has relatively high accumulation rates and thus a good position to date and compare annually resolved data with both instrumental- and model output data. Therefore, the data presented in the paper are important and will add an important piece to these estimates.
In addition to establishing the temporal variability of the snow accumulation, this study aims towards reducing uncertainties in density and strain thinning, something that becomes an issue when reconstructing accumulation at depth. I consider the method description thoroughly handled and think this is one of the strengths of the paper.
Even though the MBS record is a valuable addition to the information from the existing ice cores I don’t consider the results groundbreaking (e.g. line 203-205). The great variability of spatially and temporally is precisely why more records are necessary to get a better idea about the SMB of Antarctica.
There are some questions and concerns that I would like the authors to clarify before the paper is ready for publication.
Major comments
1. Accumulation is not necessarily equal to snowfall precipitation. SMB is the net result of multiple processes such as surface mass gains from snowfall and deposition and losses from snow sublimation and wind erosion. There is a very brief meeting of the meteorological conditions in section 2.1 where previous studies of snow accumulation are mentioned. As I understand, the sites are on the escarpment between the coastal area and the polar plateau, and the wind is likely an important factor. However, the wind direction is mentioned but nothing about the wind speed. Maybe the meteorological conditions are discussed in the previous papers but are fundamental when discussing snow accumulation and must be included in some form also here. I also wonder if there is any contribution of dry deposition?
2. Much of the discussion is based on having established a solid annual resolved chronology. However, not much is presented in the paper but refers to previous papers. I think that the dating error is important to highlight. There are several places in the discussion when specific years are mentioned and/or very brief time spans. To facilitate the (critical) reader I would suggest including a summary of the dating from Vance et al., 2024.
3. Accumulation is closely linked to climate and when discussing the temporal variability on larger scale as is done here, it would be relevant to present the d18O record. For instance, in section 4.2 a potential connection between volcanic eruptions and low snow accumulation is mentioned. Looking at Fig 3a it seems like there is no major eruption in connection with the higher accumulation regimes. I think there are modelling suggesting that volcanic eruptions influence global precipitation. I understand that this is a purely speculative discussion in the connect of this paper but nevertheless interesting to think how the injection of aerosol at low latitudes could affect the transport of moisture to Antarctica.
4. I think it also important to include more about the site specifics when discussing spatial variability. I have mentioned the meteorological conditions, but what about radar to resolve the spatial variability of snow layers at different depths?
I suggest to read Eisen et al. 2008. Ground-based measurements of spatial and temporal variability of snow accumulation in East Antarctica, Rev. Geophys., 46, RG2001, http://dx.doi.org/10.1029/2006RG000218 11 April 2008
5. The paper reads well but when I came to Section 4.1. I am struggling. One problem is that there are many numbers in the text. Here it is important to include CE when referring to years. There are also some unnecessarily long sentences. I also suggest abbreviating ice equivalent throughout the paper.
Specific and technical comments
Please include CE when years are mentioned. It is not consistent throughout the paper. It sometimes gets confusing when the are several sets of numbers in the same sentence.
Fig. 1. DSS needs to be explained in the captions.
Line 58. Previous studies… based on what method?
Line 65. As mentioned above – a summary of chronology development is in place here.
Line 68-73. I found this section hard to follow. If I understand it correctly it might not even be necessary to include?
Line 174. “The change point in 1461...”. Here is one example of where it is necessary to include CE. Probably rather use “around” and include the dating error.
Line 256. Maybe the expression “basin-wide representation” is somewhat optimistic…
Citation: https://doi.org/10.5194/egusphere-2026-4880-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 263 | 146 | 65 | 474 | 58 | 72 | 59 |
- HTML: 263
- PDF: 146
- XML: 65
- Total: 474
- Supplement: 58
- BibTeX: 72
- EndNote: 59
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of
High centennial-scale variability in an East Antarctic snow accumulation record spanning the last millennium
by Nilssen and others
General
The paper presents a new millennial-length annually-resolved snow accumulation record from Wilkes Land, East Antarctica. Such long accumulation time series data are valuable: they provide important long-term context for observed temporal surface accumulation variability based on (much) shorter time series from e.g. stakes, shallow firn cores and/or atmospheric models. The results presented in this study show significant regime shifts that appear to differ significantly from the (relatively) nearby deep ice core drilled at the coastal site Law Dome. Interesting information is also obtained from using two different ways to correct for strain induced layer thinning. The paper does lack some contextualization and remains a little bit thin on interpretation, so that sometimes it reads more like a data paper. But it is worth publishing as the new time series is important and the analysis robust. Furthermore, the paper is well written (pending some clarifications, see below) and the figures are of good quality. My comments are therefore relatively minor.
Major comments
When discussing the drill site, it would be nice to provide some climatological context, e.g. you could provide maps encompassing the same area as in Fig. 1 to present and briefly discuss annual average temperature, annual accumulation rate (from e.g. ERA5).
Maybe I missed it, but can you say something about the ice velocity at this site, i.e. how far the core site travelled horizontally over the period the accumulation record covers? Compared to the regional accumulation gradients, what does this advection imply for the consistency of the accumulation time series in the core (probably little, but good to state an estimate)?
Figure 2: Can you comment on the fact that the empirical fit appears to systematically overestimate density below a depth of ~150 m?
line 156: How sensitive are your results to the choice of l (here l = 50)?
Can you provide an estimate of the (relative) uncertainty in your ice core-derived annual accumulation values?
Minor and textual comments
The first sentence of the abstract is inconsistent. Please reformulate, e.g. consider replacing "ice loss at the margins of the Antarctic Ice Sheet" with "ice loss at the margins of the West Antarctic Ice Sheet". Or replace Mass balance" by "Surface mass balance"?
line 3: "snowfall variability" do you mean "snow accumulation variability"?
line 25: "Surface mass balance, which is dominated by snowfall variability" suggest "Surface mass balance variability, which is dominated by snowfall variability".
line 28: "Ice core records extend the climate record" suggest "Ice core records extend the record of direct accumulation observations". Because ice cores are part of the climate record.
line 61 and elsewhere: surface -> shallow (suggestion)
line 68: Please clarify "unconstrained" in this context.
line 147: Why "thus"?
line 274: snowfall accumulation records -> snowfall accumulation records derived from ice cores